For risk managers, underwriters and compliance leads deciding whether to trust an AI agent with compliance work
AI Agent Compliance Workflow Tester gives you end-to-end compliance workflows to run your own AI agent through, and scores its answer step by step: completed, derailed at a named step, or failed, with the rule behind every step.
Scoring happens in this browser: nothing you paste leaves it unless you save the run.
Each scenario has a free pack: a brief, the inputs (every one marked SPECIMEN) and an answer template. Give your agent the pack, never the scenario page: the page is the answer key.
In your own environment, with your own prompt and tools. Ask it to print its answer between two marker lines. Run it ten times if you like; agents are not deterministic.
Paste the answer or the whole transcript. Each step gets four checks (rule, decision, record, actions), the first failing step decides the result, and the sheet carries a fingerprint anyone can re-check.
Vendor onboarding with a bank detail change, a phished mailbox under the Privacy Act, an NDIS worker barred mid-shift, a ransomware payment report, an APRA insurer's provider incident, an AI claims agent's impact assessment, a material service provider, and an auditor's access review request. Filter them by process, framework, jurisdiction and risk.
A franchise head office onboards a cloud vendor that will hold franchisee identity documents and bank details.
A small physiotherapy group finds a phished mailbox full of client referral letters was read by an attacker.
A registered NDIS provider learns a labour-hire support worker is under an interim bar, a day after an alleged assault.
A freight company's negotiator pays a ransom on its behalf; the 72-hour report clock is running.
An insurer's claims platform provider is breached and the platform stays down past its tolerance.
An insurer wants an AI agent to approve small home contents claims without review.
An insurer plans to outsource claims handling to a provider whose team works overseas.
A service auditor asks for the quarter's access review and proof every leaver lost access on time.
A cyber or technology underwriter asking how an insured tests its agents, a model risk committee validating an agent before deployment, a franchise head office deciding whether an agent may onboard suppliers: each wants a record of what was tested, on which version, with which result. The score sheet is that record: the scenario and version, the agent as you entered it, every step with its four checks, the rule behind each step and the reason the first failing step failed.
Each public scenario also has held-out variants on the paid plans: one fact changed so the expected decision changes, kept out of every public file, so an agent tuned to the public answer key shows it. How scoring works.
Every scenario page, every agent pack, the scorer (one run, many runs, compare, re-verify, evidence) and the specimen runs are free. Solo USD 99 a month saves runs and adds the held-out variants, the full pack and the committee summary; Team USD 299 adds five people and the change log; Underwriter USD 990 keeps up to 25 insureds or clients apart. Pricing.